VMD and self-attention mechanism-based Bi-LSTM model for fault detection of optical fiber composite submarine cables

Abstract As the main electrical equipment of offshore power grids, optical fiber composite submarine cables undertake task transmission and data communication. In order to ensure proper functioning cable, it is necessary analyze working state identify fault event. This paper proposes a detection method for cables, that is, VMD self-attention-based Bi-LSTM model. First, we use ANSYS software generate vibration waveforms three events cables. Then, by generating matrix background noise waveforms, can realize orientation in single cable. addition, signal be decomposed into IMF components using variational mode decomposition (VMD) feature extraction. Moreover, are input self-attention layer fusion module further Finally, result output through classification layer. According comparative experiment ablation experiment, proposed model has proved outperform other benchmark models robust stable under condition different signal-to-noise ratios.